No. Sorry but the most hated, but best answer here is: It depends.... On that your data looks like. A good thing I use to do is just plot the entire series and look at it to try to identify the patterns myself. Then look at how many data points your identified (lowest frequency) pattern takes, and try to get some (min. 2/3) of them in your swarm. So if you see the longest patterns having ~2000 data points, try to swarm over >6000 :)
That way you'll get the best model Parameters for all the sequences in your data :) Best, Pascal ____________________________ BE THE CHANGE YOU WANT TO SEE IN THE WORLD ... PLEASE NOTE: This email and any file transmitted are strictly confidential and/or legally privileged and intended only for the person(s) directly addressed. If you are not the intended recipient, any use, copying, transmission, distribution, or other forms of dissemination is strictly prohibited. If you have received this email in error, please notify the sender immediately and permanently delete the email and files, if any. Please consider the environment before printing this message. > On 23 Nov 2015, at 13:32, Wakan Tanka <[email protected]> wrote: > >> On 11/22/2015 10:49 PM, Raf wrote: >> Hi guys! >> >> Just a newbie question... >> >> I guess the bigger the dataset the more examples are analyzed and the >> more accurate will be the model generated by the swarming process. >> Unfortunately, though, I noticed quite a clear increase in its duration >> directly correlated to the datasets' size. What's a good rule of thumb? >> How big should the dataset be in order to generate a good enough model >> without wasting resources and time? >> >> >> >> Thanks a lot :) > > Hello Raf, > > I was told once that 3000 line should be enough. > > Wakan >
